2 / 2024-02-25 10:17:47

Feature extraction method of rolling bearing fault based on VMD optimized by enhanced SSA and envelope analysis

rolling bearing,feature extraction,variational mode decomposition,sparrow search algorithm

The vibration signals of rolling bearing fault represent nonlinear and nonstationary characteristics with strong background noise, so as to difficultly extract its characteristic frequency and diagnose. To overcome the problem, a feature extraction method based on variational mode decomposition (VMD) optimized by enhanced sparrow search algorithm (ESSA) and envelope analysis is proposed in this paper. In the part of signal pre-processing model, a novel ESSA algorithm is proposed to optimize the parameters in the of VMD, which adopted the joint metric combined by mutual information entropy (MIE) and envelope entropy (EE) as fitness function. The signal pre-processing model separated the noised signal into a series of intrinsic mode functions (IMFs). The best IMF which contained the most fault information was selected by lowest EE value among all IMFs. Subsequently, the feature extraction was constructed via the selected IMF and envelope analysis in this study; and the fault characteristics in the envelope spectrum was extracted to recognize the real faulty conditions. To verify the function and effectivity of the proposed method, the measured signals of bearing fault was implemented. The experimental results demonstrated that the proposed method represented the fault frequency more explicitly than other comparative methods under the background of strong noise.